Bluetooth Headphone/Headset Survey Review

This story originally appeared at

Please don't consider this a comprehensive overview of the best wireless Bluetooth headphones out isn't. Manufacturers and PR people keep offering to send BT headphones to me, I keep being interested in the category, but it wasn't until a had some significant experience with BT headphones that I began to feel like I could comment on the relative performance of BT headphones I had heard. Well, with a dozen or so on hand and mounting experience, and the shopping season upon us, I figured it was time to do something.

I rounded up all the models I had, charged them all up, took a fresh listen, and began to organize my thoughts about the group. A couple fell out due to failures (the Marley Legend had a BT channel out, but sounded pretty darn good; may review later) or going out of production. And a couple were impossible to measure (Beats PowerBeats, sound was okay, though; Phiaton MS530 w/ANC on).

Another thing I had to do was take wireless measurements. Turns out, this is a bit tricky. For starters, I had to modify the testing program and spreadsheet template. For example, impedance and electrical phase measurements just aren't relevant with wireless cans.

The I ran into a little problem with impulse response plots. The Audio Precision tester uses an MLS signal to calculate impulse response and, as I understand it, the machine does a tricky thing called "autocorrelation" where the signal going to the headphones is compared against the signal coming out of the head in a rather complex way resulting in an impulse response plot. It seems the latency going through the BT transmitter cause a loss of synchronization between the outgoing and incoming MLS signal, and the tester fails to produce a plot. Oh well, no impulse respons plots for BT headphones. Fortunately, I can still measure the 300Hz square wave, and that will give you a good look at the time domain performance of the headphones.

The other thing that's a bit of a buggaboo is that there is now another active component in the system: the Bluetooth transmitter. I did a little research, and the lowest latency aptX transmitter I found was the Avantree Priva. I turn the headphone volume up all the way when testing BT headphones, but the loudness of the various headphones for a fixed input signal differs from unit to unit. That means when the testing program regulates the loudness in the headphones for each test (usually 90dBspl) the voltage needed to drive the BT transmitter differs. It's quite likely with this low-cost device that it's measured THD+noise performance with varying input levels will change, and that change is likely to show up on the headphone measurements. So, as if you don't need to take headphone measurements with a great big grain of salt already, you'll need to be doubly skeptical of BT measurements—primarily the THD+noise.

Lastly, I'll note that very few of these BT headphones perform anywhere near as well as similar wired versions. I find this puzzling. aptX BT is a pretty good quality transport mechanism, and headphone these days are quite efficient, there really shouldn't be too much trouble making a BT headphone sound near as good as a similar headphone wired out of a phone. The skeptic in me thinks they design passive cans with sound in mind, but when manufacturers get to the BT headphone they put the electronics in the earpiece and call it good without really trying to optimize the acoustics.

Alrightythen, on with the show...